--------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  C:\Users\bauschaw\Documents\awb257 (POLFS1)\experiment2\data\psrm.log
  log type:  text
 opened on:  12 Oct 2015, 10:06:11

. 
. clear 

. use "C:\Users\bauschaw\Documents\awb257 (POLFS1)\experiment2\data\psrm.dta", clear

. 
. xtset uniquesubject period
       panel variable:  uniquesubject (unbalanced)
        time variable:  period, 1 to 20
                delta:  1 unit

. gen log_period = ln(period)

. gen log_timesleader = ln(timesleader)

. gen democracy = 0

. replace democracy = 1 if votesneeded == 3
(1176 real changes made)

. 
. 
. 
. gen myDist = dist1 if playernum == 1
(1960 missing values generated)

. replace myDist = dist2 if playernum == 2
(392 real changes made)

. replace myDist = dist3 if playernum == 3
(392 real changes made)

. replace myDist = dist4 if playernum == 4
(392 real changes made)

. replace myDist = dist5 if playernum == 5
(392 real changes made)

. replace myDist = dist6 if playernum == 6
(392 real changes made)

. 
. gen totaluse = dist1 + dist2 + dist3 + dist4 + dist5 + dist6 + pubgood

. gen useZero = 0

. replace useZero =1  if totaluse == 0 & finalgrouppoints != 0 
(148 real changes made)

.         
. 
. ** This is only correct for the leader
. gen privgood = dist1 + dist2 + dist3 + dist4 + dist5 + dist6 - myDist if leader == 1
(1960 missing values generated)

. 
. 
. ** endowment dummies
. gen endow200 = 0

. replace endow200 = 1 if grouproundpoints ==200
(564 real changes made)

. gen endow150 = 0

. replace endow150 = 1 if grouproundpoints == 150
(1224 real changes made)

. 
. gen nowar =0

. replace nowar =1 if accept1 == 1 & oppresponse1 == 1
(1740 real changes made)

.         
.         
. ***** PSRM paper
. *** Table 1
. * Produces Table 1, Model 1
. xtlogit vote roundpayoff i.democracy i.grouproundpoints  log_timesleader if leader =
> = 0 & nowar==1

Fitting comparison model:

Iteration 0:   log likelihood = -959.39431  
Iteration 1:   log likelihood = -712.20852  
Iteration 2:   log likelihood = -710.31328  
Iteration 3:   log likelihood = -710.30686  
Iteration 4:   log likelihood = -710.30686  

Fitting full model:

tau =  0.0     log likelihood = -710.30686
tau =  0.1     log likelihood = -676.35761
tau =  0.2     log likelihood = -651.56672
tau =  0.3     log likelihood = -633.14469
tau =  0.4     log likelihood = -619.36057
tau =  0.5     log likelihood = -609.30228
tau =  0.6     log likelihood = -602.70649
tau =  0.7     log likelihood =  -600.0254
tau =  0.8     log likelihood = -603.11445

Iteration 0:   log likelihood = -599.88961  
Iteration 1:   log likelihood = -593.48578  
Iteration 2:   log likelihood = -593.45873  
Iteration 3:   log likelihood = -593.45872  

Random-effects logistic regression              Number of obs      =      1450
Group variable: uniquesubject                   Number of groups   =       114

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =      12.7
                                                               max =        19

                                                Wald chi2(5)       =    208.81
Log likelihood  = -593.45872                    Prob > chi2        =    0.0000

----------------------------------------------------------------------------------
            vote |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     roundpayoff |   .0511349   .0040649    12.58   0.000     .0431678    .0591019
     1.democracy |   .6431799   .3719019     1.73   0.084    -.0857344    1.372094
                 |
grouproundpoints |
            150  |  -.3937417   .2171099    -1.81   0.070    -.8192693    .0317859
            200  |  -1.397755   .3065662    -4.56   0.000    -1.998614   -.7968961
                 |
 log_timesleader |  -.1171892   .1161301    -1.01   0.313    -.3448001    .1104217
           _cons |  -2.532258   .3499797    -7.24   0.000    -3.218205    -1.84631
-----------------+----------------------------------------------------------------
        /lnsig2u |   1.060947   .2083979                      .6524947      1.4694
-----------------+----------------------------------------------------------------
         sigma_u |   1.699737   .1771109                      1.385758    2.084856
             rho |   .4675705   .0518803                      .3685709    .5691912
----------------------------------------------------------------------------------
Likelihood-ratio test of rho=0: chibar2(01) =   233.70 Prob >= chibar2 = 0.000

. est store A

. 
. 
. * Produces Table 1, Model 2 
. xttobit roundpayoff i.democracy i.grouproundpoints log_timesleader if leader == 0 & 
> nowar==1, ll(0) 

Obtaining starting values for full model:

Iteration 0:   log likelihood =  -6808.508
Iteration 1:   log likelihood = -6802.6318
Iteration 2:   log likelihood = -6802.5951
Iteration 3:   log likelihood = -6802.5951

Fitting full model:

Iteration 0:   log likelihood =  -6427.971  
Iteration 1:   log likelihood = -6414.1127  
Iteration 2:   log likelihood = -6413.9999  
Iteration 3:   log likelihood = -6413.9998  

Random-effects tobit regression                 Number of obs      =      1450
Group variable: uniquesubject                   Number of groups   =       114

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =      12.7
                                                               max =        19

                                                Wald chi2(4)       =    636.38
Log likelihood  = -6413.9998                    Prob > chi2        =    0.0000

----------------------------------------------------------------------------------
     roundpayoff |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     1.democracy |    15.9186   2.473734     6.44   0.000     11.07017    20.76703
                 |
grouproundpoints |
            150  |   22.66885   1.958387    11.58   0.000     18.83048    26.50722
            200  |   37.13883   2.398679    15.48   0.000      32.4375    41.84015
                 |
 log_timesleader |   18.66116   .9838298    18.97   0.000     16.73289    20.58943
           _cons |    12.8115   2.661457     4.81   0.000     7.595144    18.02786
-----------------+----------------------------------------------------------------
        /sigma_u |   10.23437    1.34129     7.63   0.000     7.605492    12.86325
        /sigma_e |   28.00772   .5985804    46.79   0.000     26.83453    29.18092
-----------------+----------------------------------------------------------------
             rho |   .1177972   .0282988                      .0711894    .1826926
----------------------------------------------------------------------------------

  Observation summary:       152  left-censored observations
                            1298     uncensored observations
                               0 right-censored observations

. est store B

. * Produces predictions from text on bottom of page 13
. margins democracy, predict(ystar(0,.)) post

Predictive margins                                Number of obs   =       1450
Model VCE    : OIM

Expression   : E(roundpayoff*|roundpayoff>0), predict(ystar(0,.))

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
   democracy |
          0  |   61.08687   1.658046    36.84   0.000     57.83716    64.33658
          1  |    76.4559   1.729827    44.20   0.000      73.0655     79.8463
------------------------------------------------------------------------------

. test 0.democracy == 1.democracy

 ( 1)  0bn.democracy - 1.democracy = 0

           chi2(  1) =   41.64
         Prob > chi2 =    0.0000

. 
. * Produces Table 1, Model 3 
. xttobit pubgood i.democracy i.grouproundpoints log_timesleader if leader == 1 & useZ
> ero ==0 & nowar==1, ll(0) ul(finalgrouppoints)

Obtaining starting values for full model:

Iteration 0:   log likelihood =  -1456.247
Iteration 1:   log likelihood = -1446.4057
Iteration 2:   log likelihood = -1443.4871
Iteration 3:   log likelihood = -1443.3639
Iteration 4:   log likelihood = -1443.3635

Fitting full model:

Iteration 0:   log likelihood = -930.24639  
Iteration 1:   log likelihood = -888.65324  
Iteration 2:   log likelihood = -886.74132  
Iteration 3:   log likelihood = -886.73735  
Iteration 4:   log likelihood = -886.73735  

Random-effects tobit regression                 Number of obs      =       271
Group variable: uniquesubject                   Number of groups   =        45

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.0
                                                               max =        17

                                                Wald chi2(4)       =    112.38
Log likelihood  = -886.73735                    Prob > chi2        =    0.0000

----------------------------------------------------------------------------------
         pubgood |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     1.democracy |   67.67872    22.9775     2.95   0.003     22.64364    112.7138
                 |
grouproundpoints |
            150  |   106.7564   12.94272     8.25   0.000     81.38913    132.1237
            200  |   132.3339   15.70143     8.43   0.000     101.5597    163.1081
                 |
 log_timesleader |   33.88043   7.448747     4.55   0.000     19.28116    48.47971
           _cons |  -10.10291   20.05733    -0.50   0.614    -49.41455    29.20874
-----------------+----------------------------------------------------------------
        /sigma_u |   62.47382   10.87432     5.75   0.000     41.16055    83.78709
        /sigma_e |   68.36768   4.817522    14.19   0.000     58.92551    77.80985
-----------------+----------------------------------------------------------------
             rho |   .4550455    .094071                       .281618    .6376536
----------------------------------------------------------------------------------

  Observation summary:        15  left-censored observations
                             140     uncensored observations
                             116 right-censored observations

. est store C

. * Produces Figure 1
. margins democracy, at(grouproundpoints = (100 (50) 200)) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        271
Model VCE    : OIM

Expression   : E(pubgood*|pubgood>0), predict(ystar(0,.))

1._at        : grouproundpoints=         100

2._at        : grouproundpoints=         150

3._at        : grouproundpoints=         200

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |    62.8986   12.71093     4.95   0.000     37.98564    87.81156
         1 1  |   115.1868   16.21782     7.10   0.000      83.4005    146.9732
         2 0  |   150.4601   16.52888     9.10   0.000      118.064    182.8561
         2 1  |   215.7781   17.72495    12.17   0.000     181.0379    250.5184
         3 0  |   174.7186   19.39691     9.01   0.000     136.7014    212.7359
         3 1  |   241.0944   19.98994    12.06   0.000     201.9148    280.2739
-------------------------------------------------------------------------------

. marginsplot, legend(off) xtitle("Endowment") ytitle("Public Good Investment") title(
> "") ///
>  plot1opts(msymbol(D)) plot2opts(msymbol(S)) xscale(range(90 210))

  Variables that uniquely identify margins: grouproundpoints democracy

. test 0.democracy#1._at == 1.democracy#1._at 

 ( 1)  1bn._at#0bn.democracy - 1bn._at#1.democracy = 0

           chi2(  1) =    8.83
         Prob > chi2 =    0.0030

. test 0.democracy#2._at == 1.democracy#2._at 

 ( 1)  2._at#0bn.democracy - 2._at#1.democracy = 0

           chi2(  1) =    8.82
         Prob > chi2 =    0.0030

. test 0.democracy#3._at == 1.democracy#3._at

 ( 1)  3._at#0bn.democracy - 3._at#1.democracy = 0

           chi2(  1) =    8.79
         Prob > chi2 =    0.0030

. 
. 
. * Produces Table 1, Model 4
. xttobit privgood i.democracy i.grouproundpoints log_timesleader if leader == 1 & use
> Zero ==0 & nowar==1, ll(0) ul(finalgrouppoints)

Obtaining starting values for full model:

Iteration 0:   log likelihood =  -1207.504
Iteration 1:   log likelihood = -1202.3408
Iteration 2:   log likelihood = -1201.9836
Iteration 3:   log likelihood = -1201.9815
Iteration 4:   log likelihood = -1201.9815

Fitting full model:

Iteration 0:   log likelihood = -398.01523  
Iteration 1:   log likelihood = -294.51632  (not concave)
Iteration 2:   log likelihood = -262.78467  
Iteration 3:   log likelihood = -249.73104  
Iteration 4:   log likelihood = -249.26865  
Iteration 5:   log likelihood = -249.25795  
Iteration 6:   log likelihood = -249.25792  
Iteration 7:   log likelihood = -249.25792  

Random-effects tobit regression                 Number of obs      =       271
Group variable: uniquesubject                   Number of groups   =        45

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.0
                                                               max =        17

                                                Wald chi2(4)       =      9.39
Log likelihood  = -249.25792                    Prob > chi2        =    0.0520

----------------------------------------------------------------------------------
        privgood |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     1.democracy |  -87.03696    33.1099    -2.63   0.009    -151.9312   -22.14275
                 |
grouproundpoints |
            150  |   -22.8938   21.91903    -1.04   0.296    -65.85432    20.06672
            200  |  -17.59131    27.0704    -0.65   0.516    -70.64832     35.4657
                 |
 log_timesleader |  -22.65431   14.49216    -1.56   0.118    -51.05842    5.749805
           _cons |  -10.37665   29.50319    -0.35   0.725    -68.20184    47.44853
-----------------+----------------------------------------------------------------
        /sigma_u |   70.08577   20.76286     3.38   0.001      29.3913    110.7802
        /sigma_e |   69.24532    10.5975     6.53   0.000      48.4746    90.01604
-----------------+----------------------------------------------------------------
             rho |   .5060318   .1594828                      .2210972    .7877774
----------------------------------------------------------------------------------

  Observation summary:       236  left-censored observations
                              35     uncensored observations
                               0 right-censored observations

. est store D

. * Produces Figure 2
. margins democracy, at(grouproundpoints = (100 (50) 200)) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        271
Model VCE    : OIM

Expression   : E(privgood*|privgood>0), predict(ystar(0,.))

1._at        : grouproundpoints=         100

2._at        : grouproundpoints=         150

3._at        : grouproundpoints=         200

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |   21.84736   9.137867     2.39   0.017     3.937471    39.75725
         1 1  |   4.555452    2.95918     1.54   0.124    -1.244434    10.35534
         2 0  |   15.20903   6.299031     2.41   0.016     2.863158    27.55491
         2 1  |   2.755191   1.899238     1.45   0.147    -.9672471     6.47763
         3 0  |   16.59034    8.33839     1.99   0.047     .2473936    32.93328
         3 1  |   3.106464   2.433749     1.28   0.202    -1.663596    7.876525
-------------------------------------------------------------------------------

. marginsplot, legend(off) xtitle("Endowment") ytitle("Private Goods") title("") ///
>  plot1opts(msymbol(D)) plot2opts(msymbol(S)) xscale(range(90 210))

  Variables that uniquely identify margins: grouproundpoints democracy

. test 0.democracy#1._at == 1.democracy#1._at 

 ( 1)  1bn._at#0bn.democracy - 1bn._at#1.democracy = 0

           chi2(  1) =    4.41
         Prob > chi2 =    0.0358

. test 0.democracy#2._at == 1.democracy#2._at 

 ( 1)  2._at#0bn.democracy - 2._at#1.democracy = 0

           chi2(  1) =    4.57
         Prob > chi2 =    0.0325

. test 0.democracy#3._at == 1.democracy#3._at

 ( 1)  3._at#0bn.democracy - 3._at#1.democracy = 0

           chi2(  1) =    3.48
         Prob > chi2 =    0.0622

. 
. gen privBinary = 0

. replace privBinary = 1 if privgood != 0
(2010 real changes made)

. * Produces Table 1, Model 5
. xtlogit privBinary i.democracy i.grouproundpoints log_timesleader if leader == 1 & u
> seZero ==0 & nowar==1

Fitting comparison model:

Iteration 0:   log likelihood = -104.27271  
Iteration 1:   log likelihood = -84.827456  
Iteration 2:   log likelihood = -81.094561  
Iteration 3:   log likelihood = -81.014361  
Iteration 4:   log likelihood = -81.014115  
Iteration 5:   log likelihood = -81.014115  

Fitting full model:

tau =  0.0     log likelihood = -81.014115
tau =  0.1     log likelihood = -79.329025
tau =  0.2     log likelihood = -77.754572
tau =  0.3     log likelihood = -76.287695
tau =  0.4     log likelihood = -74.929312
tau =  0.5     log likelihood = -73.688791
tau =  0.6     log likelihood = -72.591569
tau =  0.7     log likelihood = -71.700229
tau =  0.8     log likelihood = -71.261363

Iteration 0:   log likelihood = -71.715395  
Iteration 1:   log likelihood =   -68.2369  
Iteration 2:   log likelihood = -67.771965  
Iteration 3:   log likelihood = -67.693598  
Iteration 4:   log likelihood = -67.693598  (backed up)
Iteration 5:   log likelihood = -67.689642  
Iteration 6:   log likelihood = -67.689641  

Random-effects logistic regression              Number of obs      =       271
Group variable: uniquesubject                   Number of groups   =        45

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.0
                                                               max =        17

                                                Wald chi2(4)       =     10.09
Log likelihood  = -67.689641                    Prob > chi2        =    0.0389

----------------------------------------------------------------------------------
      privBinary |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     1.democracy |   -3.86538    1.40724    -2.75   0.006     -6.62352    -1.10724
                 |
grouproundpoints |
            150  |  -1.093587   .7511074    -1.46   0.145     -2.56573     .378557
            200  |  -.5252386   .8821527    -0.60   0.552    -2.254226    1.203749
                 |
 log_timesleader |  -.2502858   .4701539    -0.53   0.594     -1.17177    .6711988
           _cons |  -.0558898   1.094718    -0.05   0.959    -2.201498    2.089718
-----------------+----------------------------------------------------------------
        /lnsig2u |   2.404551   .6671126                      1.097034    3.712068
-----------------+----------------------------------------------------------------
         sigma_u |   3.327681   1.109969                      1.730685     6.39831
             rho |   .7709536   .1178015                      .4765639    .9256161
----------------------------------------------------------------------------------
Likelihood-ratio test of rho=0: chibar2(01) =    26.65 Prob >= chibar2 = 0.000

. est store E

. *margins democracy, at(grouproundpoints = (100 (50) 200)) predict(pu0) post
. *marginsplot, legend(off) xtitle("Endowment") ytitle("Private Goods") title("") ///
> * plot1opts(msymbol(D)) plot2opts(msymbol(S))
. *test 0.democracy#1._at == 1.democracy#1._at 
. *test 0.democracy#2._at == 1.democracy#2._at 
. *test 0.democracy#3._at == 1.democracy#3._at
. 
. *lqreg privBinary democracy  endow150 endow200 log_timesleader if leader == 1 & useZ
> ero ==0 & wonwar==-1, cluster(uniquesubject) quantiles(.15) 
. 
. 
. * To create Tables
. esttab A B C D E using table1.tex, replace f ///
>         label booktabs b(3) p(3) eqlabels(none) alignment(S S) collabels("\multicolu
> mn{1}{c}{$\beta$ / SE}") ///
>         star(* 0.10 ** 0.05 *** 0.01)
(output written to table1.tex)

. 
. 
. ******* Lags 
. sort uniquesubject period 

. by uniquesubject: gen lag_vote = vote[_n-1] if timesleader != 1 
(654 missing values generated)

. replace lag_vote = . if leader == 1
(283 real changes made, 283 to missing)

. 
. by uniquesubject: gen lag_roundpayoff = roundpayoff[_n-1] if timesleader != 1 
(654 missing values generated)

. replace lag_roundpayoff = . if leader == 1
(283 real changes made, 283 to missing)

. 
. by uniquesubject: gen lag_nowar = nowar[_n-1] if timesleader != 1
(654 missing values generated)

. 
. by uniquesubject: gen lag_grouproundpoints = grouproundpoints[_n-1] if timesleader !
> = 1
(654 missing values generated)

. 
. gen endowdiff = grouproundpoints - lag_grouproundpoints
(654 missing values generated)

. 
. gen myDistBinary = 0

. replace myDistBinary = 1 if myDist > 0
(311 real changes made)

. 
. 
. *** Table 2
. 
. 
. * Produces Table 2, Model 1
. xttobit roundpayoff  i.democracy##i.lag_vote endowdiff log_timesleader  if leader ==
> 0 &  useZero == 0 & nowar==1 & lag_nowar==1, ll(0) 

Obtaining starting values for full model:

Iteration 0:   log likelihood = -4146.0169
Iteration 1:   log likelihood = -4141.6904
Iteration 2:   log likelihood = -4141.6181
Iteration 3:   log likelihood =  -4141.618

Fitting full model:

Iteration 0:   log likelihood =  -4056.834  
Iteration 1:   log likelihood = -4055.0719  
Iteration 2:   log likelihood = -4055.0714  
Iteration 3:   log likelihood = -4055.0714  

Random-effects tobit regression                 Number of obs      =       905
Group variable: uniquesubject                   Number of groups   =       107

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       8.5
                                                               max =        16

                                                Wald chi2(5)       =    354.43
Log likelihood  = -4055.0714                    Prob > chi2        =    0.0000

------------------------------------------------------------------------------------
       roundpayoff |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
       1.democracy |   22.92045   4.212446     5.44   0.000      14.6642    31.17669
        1.lag_vote |   13.30094   2.918139     4.56   0.000     7.581496    19.02039
                   |
democracy#lag_vote |
              1 1  |  -10.98182   4.542318    -2.42   0.016     -19.8846   -2.079038
                   |
         endowdiff |   .2723161    .016878    16.13   0.000     .2392358    .3053965
   log_timesleader |   10.02928   1.456756     6.88   0.000     7.174093    12.88447
             _cons |   43.89937   3.586267    12.24   0.000     36.87041    50.92832
-------------------+----------------------------------------------------------------
          /sigma_u |   9.319474   1.406211     6.63   0.000     6.563351     12.0756
          /sigma_e |    23.3817   .6193226    37.75   0.000     22.16785    24.59555
-------------------+----------------------------------------------------------------
               rho |   .1370874    .037631                      .0764087    .2244209
------------------------------------------------------------------------------------

  Observation summary:        41  left-censored observations
                             864     uncensored observations
                               0 right-censored observations

. est store T

. * Produces Table 3, Left
. margins democracy,  at (lag_vote == (0 1) ) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        905
Model VCE    : OIM

Expression   : E(roundpayoff*|roundpayoff>0), predict(ystar(0,.))

1._at        : lag_vote        =           0

2._at        : lag_vote        =           1

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |   62.39993   2.467916    25.28   0.000     57.56291    67.23696
         1 1  |   85.15354   3.382496    25.17   0.000     78.52397    91.78311
         2 0  |   75.56596   2.093083    36.10   0.000     71.46359    79.66832
         2 1  |    87.4687   1.857546    47.09   0.000     83.82798    91.10942
-------------------------------------------------------------------------------

. test 0.democracy#1._at == 0.democracy#2._at 

 ( 1)  1bn._at#0bn.democracy - 2._at#0bn.democracy = 0

           chi2(  1) =   20.86
         Prob > chi2 =    0.0000

. test 1.democracy#1._at == 1.democracy#2._at 

 ( 1)  1bn._at#1.democracy - 2._at#1.democracy = 0

           chi2(  1) =    0.45
         Prob > chi2 =    0.5034

. 
. * Produces Table 2, Model 2
. xttobit  myDist i.democracy##i.lag_vote endowdiff log_timesleader  if leader ==0 & u
> seZero == 0 & nowar==1 & lag_nowar==1, ll(0) ul(finalgrouppoints)

Obtaining starting values for full model:

Iteration 0:   log likelihood = -2836.2043
Iteration 1:   log likelihood = -2829.9575
Iteration 2:   log likelihood = -2829.8113
Iteration 3:   log likelihood =  -2829.811

Fitting full model:

Iteration 0:   log likelihood = -1007.1946  (not concave)
Iteration 1:   log likelihood = -585.05989  (not concave)
Iteration 2:   log likelihood = -441.80775  
Iteration 3:   log likelihood = -324.28068  
Iteration 4:   log likelihood =  -315.1372  
Iteration 5:   log likelihood = -312.31022  
Iteration 6:   log likelihood = -312.28863  
Iteration 7:   log likelihood = -312.28749  
Iteration 8:   log likelihood = -312.28733  
Iteration 9:   log likelihood =  -312.2873  
Iteration 10:  log likelihood = -312.28729  

Random-effects tobit regression                 Number of obs      =       905
Group variable: uniquesubject                   Number of groups   =       107

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       8.5
                                                               max =        16

                                                Wald chi2(5)       =     26.69
Log likelihood  = -312.28729                    Prob > chi2        =    0.0001

------------------------------------------------------------------------------------
            myDist |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
       1.democracy |   -162.927   7065.809    -0.02   0.982    -14011.66     13685.8
        1.lag_vote |   29.22801   9.997296     2.92   0.003     9.633665    48.82235
                   |
democracy#lag_vote |
              1 1  |   110.9408   7065.798     0.02   0.987    -13737.77    13959.65
                   |
         endowdiff |  -.1326467   .0666505    -1.99   0.047    -.2632792   -.0020141
   log_timesleader |  -22.03447   6.282293    -3.51   0.000    -34.34754   -9.721399
             _cons |  -37.98331    14.8116    -2.56   0.010    -67.01352   -8.953095
-------------------+----------------------------------------------------------------
          /sigma_u |    28.9561   7.233807     4.00   0.000     14.77809     43.1341
          /sigma_e |   30.87203   4.097711     7.53   0.000     22.84067     38.9034
-------------------+----------------------------------------------------------------
               rho |   .4680088   .1299628                      .2355066    .7123554
------------------------------------------------------------------------------------

  Observation summary:       859  left-censored observations
                              46     uncensored observations
                               0 right-censored observations

. est store U

. * Produces Table 3, Center
. margins democracy,  at (lag_vote == (0 1) ) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        905
Model VCE    : OIM

Expression   : E(myDist*|myDist>0), predict(ystar(0,.))

1._at        : lag_vote        =           0

2._at        : lag_vote        =           1

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |   .7868318   .3684146     2.14   0.033     .0647524    1.508911
         1 1  |   3.70e-07   .0003338     0.00   0.999    -.0006538    .0006545
         2 0  |   3.239525    1.00922     3.21   0.001     1.261491     5.21756
         2 1  |   .2080213   .1319331     1.58   0.115    -.0505629    .4666055
-------------------------------------------------------------------------------

. test 0.democracy#1._at == 0.democracy#2._at 

 ( 1)  1bn._at#0bn.democracy - 2._at#0bn.democracy = 0

           chi2(  1) =    6.77
         Prob > chi2 =    0.0093

. test 1.democracy#1._at == 1.democracy#2._at 

 ( 1)  1bn._at#1.democracy - 2._at#1.democracy = 0

           chi2(  1) =    2.49
         Prob > chi2 =    0.1149

. 
. * Produces Table 2, Model 3
. xtprobit myDistBinary i.democracy##i.lag_vote endowdiff log_timesleader  if leader =
> =0 & useZero == 0 & nowar==1 & lag_nowar==1

Fitting comparison model:

Iteration 0:   log likelihood = -181.85812  
Iteration 1:   log likelihood = -144.61374  
Iteration 2:   log likelihood = -139.22103  
Iteration 3:   log likelihood = -138.93069  
Iteration 4:   log likelihood = -138.87871  
Iteration 5:   log likelihood = -138.86895  
Iteration 6:   log likelihood = -138.86692  
Iteration 7:   log likelihood = -138.86655  
Iteration 8:   log likelihood = -138.86648  
Iteration 9:   log likelihood = -138.86646  

Fitting full model:

rho =  0.0     log likelihood = -138.86646
rho =  0.1     log likelihood = -133.54534
rho =  0.2     log likelihood = -130.83907
rho =  0.3     log likelihood = -129.80976
rho =  0.4     log likelihood = -130.04191

Iteration 0:   log likelihood = -129.80831  
Iteration 1:   log likelihood = -123.75076  
Iteration 2:   log likelihood = -122.86072  
Iteration 3:   log likelihood = -122.83623  
Iteration 4:   log likelihood = -122.83623  (backed up)
Iteration 5:   log likelihood = -122.83611  
Iteration 6:   log likelihood =  -122.8361  

Random-effects probit regression                Number of obs      =       905
Group variable: uniquesubject                   Number of groups   =       107

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       8.5
                                                               max =        16

                                                Wald chi2(5)       =     30.02
Log likelihood  =  -122.8361                    Prob > chi2        =    0.0000

------------------------------------------------------------------------------------
      myDistBinary |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
       1.democracy |  -7.256115    1330.29    -0.01   0.996    -2614.577    2600.065
        1.lag_vote |   1.086439   .4036733     2.69   0.007     .2952544    1.877625
                   |
democracy#lag_vote |
              1 1  |   5.242887    1330.29     0.00   0.997    -2602.078    2612.564
                   |
         endowdiff |  -.0072074   .0024167    -2.98   0.003    -.0119441   -.0024708
   log_timesleader |  -.7793232   .2078921    -3.75   0.000    -1.186784   -.3718623
             _cons |  -1.467268   .5744562    -2.55   0.011    -2.593181   -.3413543
-------------------+----------------------------------------------------------------
          /lnsig2u |   .5461388   .4926793                     -.4194949    1.511772
-------------------+----------------------------------------------------------------
           sigma_u |   1.313991   .3236882                       .810789    2.129498
               rho |   .6332393   .1144234                      .3966376    .8193237
------------------------------------------------------------------------------------
Likelihood-ratio test of rho=0: chibar2(01) =    32.06 Prob >= chibar2 = 0.000

. est store V

. * Produces Table 3, Right
. margins democracy,  at (lag_vote == (0 1) ) predict(pu0) post

Predictive margins                                Number of obs   =        905
Model VCE    : OIM

Expression   : Pr(myDistBinary=1 assuming u_i=0), predict(pu0)

1._at        : lag_vote        =           0

2._at        : lag_vote        =           1

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |    .007719   .0085945     0.90   0.369     -.009126    .0245639
         1 1  |   1.19e-19   6.37e-12     0.00   1.000    -1.25e-11    1.25e-11
         2 0  |   .0650364   .0305917     2.13   0.034     .0050778     .124995
         2 1  |    .000663   .0011143     0.60   0.552    -.0015209    .0028469
-------------------------------------------------------------------------------

. test 0.democracy#1._at == 0.democracy#2._at 

 ( 1)  1bn._at#0bn.democracy - 2._at#0bn.democracy = 0

           chi2(  1) =    4.63
         Prob > chi2 =    0.0313

. test 1.democracy#1._at == 1.democracy#2._at 

 ( 1)  1bn._at#1.democracy - 2._at#1.democracy = 0

           chi2(  1) =    0.35
         Prob > chi2 =    0.5518

. 
. 
. esttab T U V using table7.tex, replace f ///
>         label booktabs b(3) p(3) eqlabels(none) alignment(S S) collabels("\multicolu
> mn{1}{c}{$\beta$ / SE}") ///
>         star(* 0.10 ** 0.05 *** 0.01)
(output written to table7.tex)

.  
.  
. *** Leader Payoffs
. *** Table 4
. gen experiencedLeader = 0 if leader == 1
(1960 missing values generated)

. replace experiencedLeader = 1 if leader ==1 & timesleader > 1
(283 real changes made)

. * Produces Table 4, Model 1
. xttobit myDist  i.democracy i.grouproundpoints if leader ==1 & useZero == 0 & wonvot
> e ==1 & nowar==1, ll(0) 

Obtaining starting values for full model:

Iteration 0:   log likelihood = -1166.8236
Iteration 1:   log likelihood = -1161.0048
Iteration 2:   log likelihood = -1160.3582
Iteration 3:   log likelihood = -1160.3308
Iteration 4:   log likelihood = -1160.3308

Fitting full model:

Iteration 0:   log likelihood = -565.10194  
Iteration 1:   log likelihood = -510.42212  
Iteration 2:   log likelihood = -508.51767  
Iteration 3:   log likelihood = -508.45985  
Iteration 4:   log likelihood = -508.45984  

Random-effects tobit regression                 Number of obs      =       237
Group variable: uniquesubject                   Number of groups   =        35

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.8
                                                               max =        17

                                                Wald chi2(3)       =     17.31
Log likelihood  = -508.45984                    Prob > chi2        =    0.0006

----------------------------------------------------------------------------------
          myDist |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
     1.democracy |  -35.89906   33.04847    -1.09   0.277    -100.6729    28.87476
                 |
grouproundpoints |
            150  |   26.24154   15.03972     1.74   0.081    -3.235767    55.71885
            200  |   69.10961   18.23642     3.79   0.000     33.36689    104.8523
                 |
           _cons |  -23.46486   25.87097    -0.91   0.364    -74.17104    27.24132
-----------------+----------------------------------------------------------------
        /sigma_u |   87.43742   16.21973     5.39   0.000     55.64733    119.2275
        /sigma_e |   49.17809   4.284695    11.48   0.000     40.78024    57.57594
-----------------+----------------------------------------------------------------
             rho |   .7596844   .0705187                      .6029563    .8748393
----------------------------------------------------------------------------------

  Observation summary:       154  left-censored observations
                              83     uncensored observations
                               0 right-censored observations

. est store W

. * Produces predictions in text on top of page 21
. margins democracy, predict(ystar(0,.)) post     

Predictive margins                                Number of obs   =        237
Model VCE    : OIM

Expression   : E(myDist*|myDist>0), predict(ystar(0,.))

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
   democracy |
          0  |   44.07946   12.07379     3.65   0.000     20.41526    67.74366
          1  |   27.77973   10.29091     2.70   0.007      7.60993    47.94954
------------------------------------------------------------------------------

. 
. * Produces Table 3, Model 2
. xttobit myDist  i.democracy##i.experiencedLeader i.grouproundpoints if leader ==1 & 
> useZero == 0 & wonvote ==1 & nowar==1, ll(0) 

Obtaining starting values for full model:

Iteration 0:   log likelihood = -1169.5119
Iteration 1:   log likelihood = -1161.0975
Iteration 2:   log likelihood = -1160.3209
Iteration 3:   log likelihood = -1160.2972
Iteration 4:   log likelihood = -1160.2971

Fitting full model:

Iteration 0:   log likelihood = -565.04213  
Iteration 1:   log likelihood = -510.16571  
Iteration 2:   log likelihood = -508.20371  
Iteration 3:   log likelihood = -508.13913  
Iteration 4:   log likelihood = -508.13911  
Iteration 5:   log likelihood = -508.13908  

Random-effects tobit regression                 Number of obs      =       237
Group variable: uniquesubject                   Number of groups   =        35

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.8
                                                               max =        17

                                                Wald chi2(5)       =     17.81
Log likelihood  = -508.13908                    Prob > chi2        =    0.0032

-----------------------------------------------------------------------------------
           myDist |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
      1.democracy |  -20.05542   39.73662    -0.50   0.614    -97.93777    57.82693
1.experiencedLe~r |    15.0303   21.32527     0.70   0.481    -26.76646    56.82706
                  |
        democracy#|
experiencedLeader |
             1 1  |  -22.69314   29.82897    -0.76   0.447    -81.15685    35.77058
                  |
 grouproundpoints |
             150  |   27.71471   15.23134     1.82   0.069    -2.138165    57.56758
             200  |   71.57499   18.61522     3.84   0.000     35.08983    108.0601
                  |
            _cons |   -35.1576    31.1537    -1.13   0.259    -96.21773    25.90252
------------------+----------------------------------------------------------------
         /sigma_u |   88.64568      16.55     5.36   0.000     56.20827    121.0831
         /sigma_e |   49.00807   4.270432    11.48   0.000     40.63817    57.37796
------------------+----------------------------------------------------------------
              rho |   .7659039   .0698078                      .6099698    .8793225
-----------------------------------------------------------------------------------

  Observation summary:       154  left-censored observations
                              83     uncensored observations
                               0 right-censored observations

. est store X

. * Produces predictions related to the text near bottom of page 21
. margins democracy,  at (experiencedLeader == (0 1)) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        237
Model VCE    : OIM

Expression   : E(myDist*|myDist>0), predict(ystar(0,.))

1._at        : experiencedLeader=           0

2._at        : experiencedLeader=           1

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |   39.33396   13.05743     3.01   0.003     13.74187    64.92606
         1 1  |    30.4032   12.65081     2.40   0.016     5.608062    55.19833
         2 0  |   47.03715   13.33373     3.53   0.000     20.90352    73.17078
         2 1  |   27.39099   10.46616     2.62   0.009     6.877701    47.90429
-------------------------------------------------------------------------------

. test 0.democracy#1._at == 0.democracy#2._at 

 ( 1)  1bn._at#0bn.democracy - 2._at#0bn.democracy = 0

           chi2(  1) =    0.51
         Prob > chi2 =    0.4738

. test 1.democracy#1._at == 1.democracy#2._at 

 ( 1)  1bn._at#1.democracy - 2._at#1.democracy = 0

           chi2(  1) =    0.13
         Prob > chi2 =    0.7204

. 
. * Produces Table 4, Model 3     
. xttobit roundpayoff  i.democracy##i.experiencedLeader i.grouproundpoints if leader =
> =1 & useZero == 0 & wonvote ==1 & nowar==1, ll(0) 

Obtaining starting values for full model:

Iteration 0:   log likelihood = -1079.0368
Iteration 1:   log likelihood = -1077.1739
Iteration 2:   log likelihood = -1077.0598
Iteration 3:   log likelihood = -1077.0587

Fitting full model:

Iteration 0:   log likelihood = -1075.2936  
Iteration 1:   log likelihood = -1075.2772  
Iteration 2:   log likelihood = -1075.2772  

Random-effects tobit regression                 Number of obs      =       237
Group variable: uniquesubject                   Number of groups   =        35

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       6.8
                                                               max =        17

                                                Wald chi2(5)       =    222.79
Log likelihood  = -1075.2772                    Prob > chi2        =    0.0000

-----------------------------------------------------------------------------------
      roundpayoff |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
      1.democracy |    12.8936   9.820526     1.31   0.189    -6.354279    32.14148
1.experiencedLe~r |   20.23344   6.345008     3.19   0.001     7.797452    32.66942
                  |
        democracy#|
experiencedLeader |
             1 1  |  -15.70806   8.883222    -1.77   0.077    -33.11885    1.702736
                  |
 grouproundpoints |
             150  |   36.33142   3.723271     9.76   0.000     29.03394    43.62889
             200  |   68.09011   4.631176    14.70   0.000     59.01317    77.16705
                  |
            _cons |   47.91571    7.45534     6.43   0.000     33.30351    62.52791
------------------+----------------------------------------------------------------
         /sigma_u |   17.88077   4.206629     4.25   0.000     9.635931    26.12561
         /sigma_e |   20.39416   1.087641    18.75   0.000     18.26242    22.52589
------------------+----------------------------------------------------------------
              rho |   .4346153   .1265438                      .2133603    .6792291
-----------------------------------------------------------------------------------

  Observation summary:         1  left-censored observation
                             236     uncensored observations
                               0 right-censored observations

. est store Y

. * Produces predictions related to text on bottom of page 21 and top of page 22
. margins democracy,  at (experiencedLeader == (0 1)) predict(ystar(0,.)) post

Predictive margins                                Number of obs   =        237
Model VCE    : OIM

Expression   : E(roundpayoff*|roundpayoff>0), predict(ystar(0,.))

1._at        : experiencedLeader=           0

2._at        : experiencedLeader=           1

-------------------------------------------------------------------------------
              |            Delta-method
              |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
_at#democracy |
         1 0  |   83.76632   6.568826    12.75   0.000     70.89166    96.64099
         1 1  |   96.60113   7.220027    13.38   0.000     82.45013    110.7521
         2 0  |   103.9286   5.160875    20.14   0.000     93.81351    114.0438
         2 1  |   101.1179    5.41673    18.67   0.000     90.50126    111.7345
-------------------------------------------------------------------------------

. test 0.democracy#1._at == 0.democracy#2._at 

 ( 1)  1bn._at#0bn.democracy - 2._at#0bn.democracy = 0

           chi2(  1) =   10.21
         Prob > chi2 =    0.0014

. test 1.democracy#1._at == 1.democracy#2._at 

 ( 1)  1bn._at#1.democracy - 2._at#1.democracy = 0

           chi2(  1) =    0.54
         Prob > chi2 =    0.4636

. 
. * Produces summary stats for footnote 17
. sum roundpayoff if  leader ==1 & useZero == 0 & wonvote ==1 & nowar==1 & democracy==
> 0 & experiencedLeader == 0

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
 roundpayoff |        17    89.03922    46.21152          0   193.6667

. sum roundpayoff if  leader ==1 & useZero == 0 & wonvote ==1 & nowar==1 & democracy==
> 0 & experiencedLeader == 1

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
 roundpayoff |       107    97.60685    33.90879   26.06667        267

. sum roundpayoff if  leader ==1 & useZero == 0 & wonvote ==1 & nowar==1 & democracy==
> 1 & experiencedLeader == 0

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
 roundpayoff |        14    92.07619    30.60881   34.06667      156.6

. sum roundpayoff if  leader ==1 & useZero == 0 & wonvote ==1 & nowar==1 & democracy==
> 1 & experiencedLeader == 1

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
 roundpayoff |        99    99.19461    25.18851   46.66667      156.6

. 
. 
.  esttab W X Y using table7.tex, replace f ///
>         label booktabs b(3) p(3) eqlabels(none) alignment(S S) collabels("\multicolu
> mn{1}{c}{$\beta$ / SE}") ///
>         star(* 0.10 ** 0.05 *** 0.01)
(output written to table7.tex)

.  
.  
.  
. ** Scatter plots
. * pub goods
. * Produces Figure 1 in the appendix
. graph twoway (scatter pubgood timesleader if democracy==0 & leader==1 & nowar==1,  m
> symbol(D) jitter(10)) ///
>         (scatter pubgood timesleader if democracy==1 & leader==1 & nowar==1, msymbol
> (S) jitter(10)) ///
>     (lowess pubgood timesleader if democracy==0 & leader==1 & nowar==1, lcolor(navy)
> ) ///
>         (lowess pubgood timesleader if democracy==1 & leader==1 & nowar==1, lcolor(m
> aroon) ///
>         xtitle("Times Leader") ytitle("Public Goods") title("") legend(label( 1 "Sma
> ll Coalition") label( 2 "Large Coalition") label( 3 "Small Coalition Lowess") label(
>  4 "Large Coalition Lowess")))

. 
. * Produces Figure 2 in the appendix
. graph twoway (scatter pubgood log_timesleader if democracy==0 & leader==1 & nowar==1
> ,  msymbol(D) jitter(10)) ///
>         (scatter pubgood log_timesleader if democracy==1 & leader==1 & nowar==1, msy
> mbol(S) jitter(10)) ///
>     (lowess pubgood log_timesleader if democracy==0 & leader==1 & nowar==1, lcolor(n
> avy)) ///
>         (lowess pubgood log_timesleader if democracy==1 & leader==1 & nowar==1, lcol
> or(maroon) ///
>         xtitle("Log of Times Leader") ytitle("Public Goods") title("") legend(label(
>  1 "Small Coalition") label( 2 "Large Coalition") label( 3 "Small Coalition Lowess")
>  label( 4 "Large Coalition Lowess")))

.                 
.         
. * priv goods    
. * Produces Figure 3 in the appendix
. graph twoway (scatter privgood timesleader if democracy==0 & leader==1 & nowar==1,  
> msymbol(D) jitter(10)) ///
>         (scatter privgood timesleader if democracy==1 & leader==1 & nowar==1, msymbo
> l(S) jitter(10)) ///
>     (lowess privgood timesleader if democracy==0 & leader==1 & nowar==1, lcolor(navy
> )) ///
>         (lowess privgood timesleader if democracy==1 & leader==1 & nowar==1, lcolor(
> maroon) ///
>         xtitle("Times Leader") ytitle("Private Goods") title("") legend(label( 1 "Sm
> all Coalition") label( 2 "Large Coalition") label( 3 "Small Coalition Lowess") label
> ( 4 "Large Coalition Lowess")))

. 
. * Produces Figure 4 in the appendix
. graph twoway (scatter privgood log_timesleader if democracy==0 & leader==1 & nowar==
> 1,  msymbol(D) jitter(10)) ///
>         (scatter privgood log_timesleader if democracy==1 & leader==1 & nowar==1, ms
> ymbol(S) jitter(10)) ///
>     (lowess privgood log_timesleader if democracy==0 & leader==1 & nowar==1, lcolor(
> navy)) ///
>         (lowess privgood log_timesleader if democracy==1 & leader==1 & nowar==1, lco
> lor(maroon) ///
>         xtitle("Log of Times Leader") ytitle("Private Goods") title("") legend(label
> ( 1 "Small Coalition") label( 2 "Large Coalition") label( 3 "Small Coalition Lowess"
> ) label( 4 "Large Coalition Lowess")))

. 
. * Lag Vote
. * Produces Figure 5 in the appendix
. graph twoway (scatter roundpayoff log_timesleader if democracy==0 & leader==0 & nowa
> r==1 & lag_vote==0 &  useZero == 0 & lag_nowar==1,  msymbol(T) jitter(10)) ///
>         (scatter roundpayoff log_timesleader if democracy==0 & leader==0 & nowar==1 
> & lag_vote==1 &  useZero == 0 & lag_nowar==1, msymbol(X) jitter(10)) ///
>         (lowess roundpayoff log_timesleader if democracy==0 & leader==0 & nowar==1 &
>  lag_vote==0 &  useZero == 0 & lag_nowar==1, lcolor(navy)) ///
>         (lowess roundpayoff log_timesleader if democracy==0 & leader==0 & nowar==1 &
>  lag_vote==1 &  useZero == 0 & lag_nowar==1, lcolor(maroon) ///
>         xtitle("Small Coalition: Log of Times Leader") ytitle("Round Payoff") title(
> "") legend(label( 1 "Lag Vote = 0") label( 2 "Lag Vote = 1") label( 3 "Lag Vote = 0 
> Lowess") label( 4 "Lag Vote = 1 Lowess")))      

. 
. * Produces Figure 6 in the appendix
. graph twoway (scatter roundpayoff log_timesleader if democracy==1 & leader==0 & nowa
> r==1 & lag_vote==0 &  useZero == 0 & lag_nowar==1,  msymbol(T) jitter(10)) ///
>         (scatter roundpayoff log_timesleader if democracy==1 & leader==0 & nowar==1 
> & lag_vote==1 &  useZero == 0 & lag_nowar==1, msymbol(X) jitter(10)) ///
>         (lowess roundpayoff log_timesleader if democracy==1 & leader==0 & nowar==1 &
>  lag_vote==0 &  useZero == 0 & lag_nowar==1, lcolor(navy)) ///
>         (lowess roundpayoff log_timesleader if democracy==1 & leader==0 & nowar==1 &
>  lag_vote==1 &  useZero == 0 & lag_nowar==1, lcolor(maroon) ///
>         xtitle("Large Coalition: Log of Times Leader") ytitle("Round Payoff") title(
> "") legend(label( 1 "Lag Vote = 0") label( 2 "Lag Vote = 1") label( 3 "Lag Vote = 0 
> Lowess") label( 4 "Lag Vote = 1 Lowess")))      

. 
. log close
      name:  <unnamed>
       log:  C:\Users\bauschaw\Documents\awb257 (POLFS1)\experiment2\data\psrm.log
  log type:  text
 closed on:  12 Oct 2015, 10:07:01
--------------------------------------------------------------------------------------
